Festival of Data Technology

In this training taster, dive into two frameworks for becoming successful with data: Value Chain of Data Science: A battle-tested framework to think about analytics solutions end-to-end. Analytics Maturity Journey: A way to assess where an organization stands today, and in what leadership should invest, to become an effective data-driven organizations.

With advances in cloud technology it's easier than ever to create a fully serverless analytics pipeline that can handle millions of events. We'll deep dive into setting up a Snowplow tracker, collector, validation and dbt models with Google Cloud Platform for €0.03/day.

A popular topic is stories from analists becoming analytics engineers, but I actually came from the other side of the spectrum. Some of my views on this path.

Data is everywhere. It is through analysis and visualization that we are able to turn data into information that can be used to drive better decision making. Out-of-the-box tools will allow you to create a chart, but if you want people to take action, your numbers need to tell a compelling story. Learn how elements of storytelling can be applied to data visualization.

Explore the issues that can come when you have imbalanced data while training a Machine Learning model. Learn the importance of choosing the right metric and how to use different resampling techniques to account for this.

From inventory to website visitors, resource planning to finance and accounting, much of any organization’s data has temporal nature. Key problems facing business include not only forecasting the future values, but also identifying when these values are alarming. Intertemporal anomaly detection often becomes a complex problem that neither common rule-based heuristics nor standard anomaly detection algorithms can tackle well. In this webinar we will have a close look at this domain, learn about its distinctive features and challenges as well as learn to detect anomalies in time series data, while dodging many common mistakes and issues.

With the democratization of data via data lakes data science teams increasingly rely on custom model pipelines for data preprocessing and feature engineering. As a result it becomes difficult to reuse features or even compare similar features across different teams. This can be a significant challenge, as it can lead to duplicative work and ambiguous definitions that cause confusion and a risk on wrong conclusions. For example, two data science teams that both communicate an average click-rate to the marketing team, where one team excludes clicks made by robots and the other doesn’t. Without being aware of this different interpretation the marketing team can make some awfully wrong decisions.

MLFlow has been widely used as a reliable ML experiment tracking and model registry to pave the way towards MLOps. However, its flexibility also imposes challenges on having a robust and consistent way working across different scenarios. In this talk we will have a closer look at how to properly use MLFlow in combination with other open source frameworks such as Airflow and Kubernetes to further mature the lifecycle of machine learning use cases.